code-transfer

Pass

Audited by Gen Agent Trust Hub on Aug 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides a local Python utility script (scripts/line_insert.py) that the agent is instructed to execute via the shell to perform precise, line-based code insertion into files within the workspace.\n- [PROMPT_INJECTION]: The skill maintains an attack surface for indirect prompt injection as it processes and writes code snippets from the agent's context to the local filesystem.\n
  • Ingestion points: Code content provided to the line_insert.py utility or via filesystem tools such as Write.\n
  • Boundary markers: None identified.\n
  • Capability inventory: The skill performs filesystem write operations, directory creation, and file backup generation.\n
  • Sanitization: The included Python script performs basic path validation to prevent directory traversal by checking for .. in file paths.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 21, 2026, 05:04 AM
Security Audit — agent-trust-hub — code-transfer